COMPUTER-AID METHOD AND DEVICE FOR PROBABILITY-BASED SPEED FORECASTING FOR VEHICLES

Description

[0001] The invention relates to a computer-aided method and a device for generating a driving cycle for a vehicle, which is suitable for simulating a driving operation, in particular a real driving operation.

[0002] Emissions guidelines for vehicles with internal combustion engines are subject to constant change, with the goal of taking into account driving conditions that increasingly approximate real-world driving conditions. One example of such emissions guidelines is the European Union's regulation of test procedures for determining vehicle emissions under real driving conditions, so-called Real Driving Emissions (RDE). These test procedures are, for example, part of the vehicle type approval process. Emissions tests can therefore no longer be conducted exclusively on a vehicle test bench with generally defined driving cycles, but must be conducted under real driving conditions to take into account, for example, the influence of real traffic conditions and the driver's actual driving behavior.

[0003] For example, driving distances at different speed ranges and minimum and maximum stopping times must be included in an RDE-compliant driving cycle, which serves as the basis for determining emissions in accordance with the guidelines. However, emissions guidelines designed to reflect real-world driving conditions on the road permit a multitude of different driving cycles, which entails enormous testing effort for vehicle manufacturers during vehicle development. To determine a vehicle's fuel consumption under real-world driving conditions, fuel consumption is typically determined for approximately 1,000 RDE-compliant driving cycles. This testing effort can be reduced by simulating a multitude of different, guideline-compliant driving cycles that take realistic driving behavior into account.

[0004] To create such driving cycles, driving cycles can be simulated using Markov chains or neural networks. However, the driving cycles generated in this way exhibit significant deviations from driving cycles measured under real road conditions. Alternatively, short driving distances measured under real road conditions can be combined in different ways to generate a driving cycle. However, driving cycles generated in this way are relatively similar to each other and thus offer insufficient variability to determine a real, average vehicle consumption. The following prior art documents relevant to the invention are: Tae-Kyung Lee et al., "Synthesis of Real-World Driving Cycles and Their Use for Estimating PHEV Energy Consumption and Charging Opportunities: Case Study for Midwest / US", IEEE Transactions on Vehicular Technology, IEEE, USA, Vol. 60, No. 9, November 1, 2011, pages 4153-4163, DOI: 10.1109 / TVT.2011.2168251; Dietrich Maximilian et al., "A Combined Markov Chain and Reinforcement Learning Approach for Powertrain-Specific Driving Cycle Generation", SAE International Journal of Advances and Current Practices in Mobility, Vol. 3, No. 1, September 15, 2020, pages 516-527, DOI: 10.4271 / 2020-01-2185.

[0005] It is an object of the present invention to generate a plurality of different driving cycles which correspond to the real driving behavior of a vehicle.

[0006] This object is achieved by a computer-aided method and device according to the independent claims. Preferred embodiments are claimed in the subclaims.

[0007] A first aspect of the invention relates to a computer-aided method for generating a driving cycle for a vehicle, which is suitable for simulating a particularly real driving operation. The computer-aided method comprises determining a state vector of the driving cycle for a current time interval from a past speed profile, providing an acceleration prediction model, determining an acceleration value taking into account probabilities resulting from the acceleration prediction model and the state vector, integrating the determined acceleration value over the current time interval to obtain a predicted speed value for a next time interval in the future, and appending the predicted speed value to the past speed profile to generate the driving cycle.

[0008] A driving cycle within the meaning of the invention is in particular a time interval to which a constant speed value is assigned, or a temporal sequence of several time intervals to which a constant speed value is each assigned.

[0009] A current time interval of a driving cycle within the meaning of the invention is in particular a time interval which immediately follows past time intervals of the driving cycle and to which a current speed value is assigned, which can have a finite value or be zero.

[0010] A past speed profile within the meaning of the invention is, in particular, a current time interval to which a speed value is assigned and / or a past time interval to which a speed value is assigned, and / or a plurality of past time intervals to which a speed value is each assigned. A speed value can have a finite value or be zero.

[0011] An acceleration value in the sense of the invention is a positive value in the case of a positive acceleration or a negative value in the case of a negative acceleration, here also called deceleration.

[0012] A state vector of a driving cycle for a current time interval within the meaning of the invention is in particular a vector whose components correspond to one or more speed values ​​and / or one or more acceleration values ​​and / or one or more values ​​of an acceleration change and / or one or more values ​​which indicate a number of time intervals.

[0013] An acceleration prediction model within the meaning of the invention is, in particular, a model for determining one or more acceleration values ​​for a current time interval or for one or more time intervals that follow the current time interval. An acceleration prediction model within the meaning of the invention can also be referred to as a conditional acceleration prediction (CAP).

[0014] The invention is based in particular on the approach that a state vector is determined from a past speed profile, i.e., at least one speed value associated with a current time interval and / or at least one past time interval, representing the current state of a driving cycle at a current time interval. Using the state vector and a probability-based acceleration prediction model, an acceleration value is determined for the current time interval. By integrating the acceleration value thus determined over the current time interval, a predicted speed value for a future time interval is obtained, which is then appended to the past speed profile.

[0015] The computer-aided method for generating a driving cycle for a vehicle according to the present invention has the advantage over the prior art that any number of driving cycles can be generated which, on the one hand, are not similar to one another in their speed profile and, on the other hand, by using the acceleration prediction model, have a high degree of similarity to driving cycles measured under real conditions.

[0016] In a preferred embodiment of the method for generating a driving cycle, determining an acceleration value taking into account probabilities resulting from the acceleration prediction model and the state vector comprises determining, by means of the acceleration prediction model as a function of the state vector, a probability value for a current scenario of acceleration and a probability value for a current scenario of deceleration and a probability value for a current scenario of a state of constant speed.Determining an acceleration value taking probabilities into account further comprises randomly selecting, for the current time interval, a scenario of acceleration, deceleration or constant speed based on the probability values ​​for current scenarios of acceleration, deceleration and constant speed and / or determining, by means of the acceleration forecast model as a function of the state vector, a probability distribution of acceleration values ​​of the randomly selected scenario and randomly selecting, for the current time interval, an acceleration value based on the probability distribution of acceleration values ​​of the randomly selected scenario.

[0017] A current scenario within the meaning of the invention is in particular an acceleration, a deceleration or a state of constant speed in a current time interval.

[0018] A random selection within the meaning of the invention is in particular a drawing of a random sample or a random drawing in the statistical sense.

[0019] Randomly selecting an acceleration, deceleration, or constant speed scenario based on probability values ​​for current acceleration, deceleration, and constant speed scenarios, as well as randomly selecting an acceleration value for the current time interval based on a probability distribution of acceleration values, has the following advantages: Starting from the same past speed history, the acceleration prediction model can be used to generate a large number of driving cycles that are not similar to one another. Furthermore, these provide sufficient variability to determine a vehicle's average fuel consumption under real-world driving conditions.

[0020] In a further preferred embodiment of the method for generating a driving cycle, the driving cycle is generated by iteratively executing the steps of the method in the listed order, and the predicted speed values ​​are each appended to the past speed curve from a previous iteration. This has the advantage that a driving cycle of any length can be generated.

[0021] In a further preferred embodiment of the method for generating a driving cycle, a plurality of predicted speed values ​​are obtained for the same time intervals in the future based on the past speed profile, so that statistical speed distributions are obtained for time intervals in the future. The statistical speed distributions for the time intervals in the future allow a statistical evaluation of the generated driving cycle. For example, a driving cycle can be created in this way whose speed values ​​correspond to the respective expected value of the statistical speed distributions.

[0022] In a further preferred embodiment of the method for generating a driving cycle, the state vector for a current time interval has at least one current speed value and / or one or more speed values ​​from the past and / or one or more acceleration values ​​of one or more time intervals and / or one or more values ​​of an acceleration change of one or more time intervals and / or a value corresponding to a number of time intervals according to the duration of a currently ongoing acceleration maneuver and / or a value corresponding to a number of time intervals according to the duration of a currently ongoing deceleration maneuver and / or a value corresponding to a number of time intervals according to the duration of a currently ongoing state of constant speed.

[0023] An acceleration change of a time interval in the sense of the invention is in particular the difference between an acceleration value within the time interval and an acceleration value in a previous time interval.

[0024] An acceleration maneuver within the meaning of the invention is, in particular, an uninterrupted acceleration process of any acceleration values ​​over one or more time intervals, which began in the past, i.e., in a past time interval before the current time interval. A currently ongoing acceleration maneuver means that the acceleration maneuver continues until the time interval immediately before the current time interval.

[0025] A deceleration maneuver within the meaning of the invention is, in particular, an uninterrupted process of decelerating any negative acceleration values ​​over one or more time intervals, which began in the past, i.e., in a past time interval before the current time interval. A currently ongoing deceleration maneuver means that the deceleration maneuver continues until the time interval immediately before the current time interval.

[0026] A constant speed state within the meaning of the invention is, in particular, the maintenance of a constant speed value over one or more time intervals, which began in the past, i.e., in a past time interval before the current time interval. A currently ongoing constant speed state means that the constant speed state continues until the time interval immediately before the current time interval.

[0027] Since the state vector of a current time interval has a value corresponding to the duration of a currently ongoing acceleration maneuver, a value corresponding to the duration of a currently ongoing deceleration maneuver and / or a value corresponding to the duration of a currently ongoing state of constant speed, the duration of acceleration maneuvers, deceleration maneuvers and / or the duration of a state of constant speed within an already generated part of a driving cycle influences the further course of the driving cycle.

[0028] Based on the actual driving behavior of a vehicle, the probability determined according to the present invention for the continuation of the acceleration maneuver in the current time interval and in future time intervals of a driving cycle is influenced by the duration of an acceleration maneuver in the past of the driving cycle. The same applies to deceleration maneuvers and constant speed conditions. The determined probability distribution of acceleration values ​​of a randomly selected scenario for a current time interval and future time intervals also depends on the duration of an acceleration maneuver, deceleration maneuver, or constant speed condition in the past of the driving cycle. This has the advantage of further increasing the similarity of the generated driving cycle to driving cycles measured under real conditions.

[0029] In a further preferred embodiment of the method for generating a driving cycle, the acceleration value, which is determined taking into account probabilities resulting from the acceleration prediction model and the state vector, is based on the duration of a currently ongoing acceleration maneuver, a currently ongoing deceleration maneuver, or a currently ongoing constant speed state. This has the advantage that the duration of an acceleration maneuver, a deceleration maneuver, or a constant speed state can be adapted to real driving conditions, thus further increasing the similarity of the generated driving cycle to driving cycles measured under real conditions.

[0030] In a further preferred embodiment of the method for generating a driving cycle, the past speed profile has at least one speed value. The at least one speed value can have a finite value or be zero. This has the advantage that a driving cycle can be generated from a single speed value.

[0031] In a further preferred embodiment of the method for generating a driving cycle, the current acceleration scenario and / or the current deceleration scenario and / or the current constant speed scenario each have a probability distribution of acceleration values. This enables a statistical evaluation of the generated driving cycle. For example, a driving cycle can be created whose speed values ​​are based on acceleration values ​​that correspond to the respective expected value of the probability distributions of acceleration values ​​in the individual time intervals.

[0032] In a further preferred embodiment of the method for generating a driving cycle, an expected value of the probability distribution of acceleration values ​​is set based on the past speed profile. The expected value of the modeled probability distribution is preferably derived based on the current speed value and a past speed value. The expected value of the modeled probability distribution is preferably set based on the current speed value and the direct temporal predecessor of the current speed value. This has the advantage that the temporal speed profile of the driving cycle has a smooth or continuous profile. Abrupt temporal jumps in the speed profile of the generated driving cycle are thus avoided, which further increases its similarity to driving cycles measured under real conditions.

[0033] In a further preferred embodiment of the method for generating a driving cycle, the acceleration prediction model is based on a statistical evaluation of measured driving data from at least one real vehicle, wherein the measured driving data from the at least one real vehicle preferably exclusively comprise a temporal sequence of speed values. The driving data of the real vehicle measured under real driving conditions are preferably used for model training, in particular for determining model parameters. This has the advantage of increasing the similarity of the driving cycles generated using the acceleration prediction model to driving cycles measured under real conditions.

[0034] A further aspect of the invention relates to a method for driving a vehicle by means of an adaptive cruise control system, in particular a driver assistance system, in particular for predictive driving functions, wherein the driving cycle of the vehicle driving ahead of the vehicle is determined by means of a computer-aided method according to one of the aforementioned embodiments.

[0035] In a preferred embodiment of the method for controlling a vehicle, the distance between the vehicle and the vehicle in front is not used as an input value or boundary condition for the adaptive cruise control, with the distance between the vehicle and the vehicle in front preferably being based on a solution of a cost function or cost optimization function. The distance between the vehicle and the vehicle in front is thus not a constant value, which has the advantage that the distance can be adapted to current traffic conditions.

[0036] In a further preferred embodiment of the method for driving a vehicle, several predicted speed values ​​are obtained based on the past speed history for the same time intervals in the future, so that statistical speed distributions are obtained for time intervals in the future. Safety conditions for driving the vehicle are derived from the statistical speed distributions. These safety conditions for driving the vehicle include, in particular, determining a minimum distance between the vehicle and the vehicle in front, which must be maintained to prevent a collision between the two vehicles. This can increase safety when driving the vehicle.

[0037] A further aspect of the invention relates to a method for generating a driving cycle for a vehicle, which is suitable for use by driver assistance systems, in particular for predictive driving functions, and has the working steps of a method for generating a driving cycle according to one of the aforementioned embodiments.

[0038] A further aspect of the invention relates to a method for analyzing at least one component of a motor vehicle, wherein the at least one component or the motor vehicle is subjected to a real or simulated test operation based on at least one driving cycle, which is determined by means of a method for generating a driving cycle according to one of the aforementioned embodiments.

[0039] Further preferably, the method for analyzing at least one component of a motor vehicle comprises checking the conformity of the plurality of predicted speed values ​​with at least one boundary condition, in particular Real Driving Emissions (RDE) guidelines, after a defined number of iterations. In particular, the checking is performed periodically after the specified number of iterations has elapsed, with the specified number of iterations corresponding to a predefined total time interval, for example, 5 minutes. This has the advantage that driving cycles compliant with the RDE guidelines can be generated.

[0040] Further preferably, the method for analyzing at least one component of a motor vehicle comprises correcting, for the current time interval, the probability value for a current acceleration scenario and / or the probability value for a current deceleration scenario and / or the probability value for a current constant speed scenario, and / or correcting the acceleration value for the current time interval based on the checking. This has the advantage that driving cycles compliant with the RDE directive can be generated.

[0041] A further aspect of the invention relates to a computer program product comprising instructions which, when executed by a computer, cause the computer to carry out the steps of a method according to one of the aforementioned embodiments.

[0042] A further aspect of the invention relates to a computer-readable medium on which a computer program product according to one of the aforementioned embodiments is stored.

[0043] A further aspect of the invention relates to a device for generating a driving cycle for a vehicle, which is suitable for simulating a, in particular real, driving operation, and has means for determining a state vector of the driving cycle for a current time interval from a past speed profile, means for providing an acceleration prediction model, means for determining an acceleration value taking into account probabilities resulting from the acceleration prediction model and the state vector, means for integrating the selected acceleration value over the current time interval in order to obtain a predicted speed value for a next time interval in the future and means for appending the predicted speed value to the past speed profile to generate the driving cycle.

[0044] A means within the meaning of the invention can be designed in hardware and / or software and in particular can have a processing unit, in particular a microprocessor unit (CPU), which is preferably connected to a memory and / or bus system in terms of data or signals, and / or one or more programs or program modules. The CPU can be designed to process instructions implemented as a program stored in a memory system, to detect input signals from a data bus and / or to output signals to a data bus. A memory system can have one or more, in particular different, storage media, in particular optical, magnetic, solid-state and / or other non-volatile media. The program can be designed in such a way that it embodies the methods described here orcapable of carrying out such a method, so that the CPU can carry out the steps of such methods and can thus in particular control and / or monitor a reciprocating piston machine.

[0045] Further features and advantages will become apparent from the following description in conjunction with the figures. The figures show, at least in part, schematically: Fig. 1 a preferred embodiment of a computer-aided method according to the invention for generating a driving cycle for a vehicle, the method being suitable for simulating real driving operation; Fig. 2 a preferred embodiment of a computer-aided method according to the invention for generating an RDE-compliant driving cycle for a vehicle; and Fig. 3 a preferred embodiment of a device for generating a driving cycle for a vehicle, which is suitable for simulating real driving operation.

[0046] Fig. 1shows a preferred embodiment of a computer-aided method 100 according to the invention for generating a driving cycle for a vehicle, wherein the method 100 is suitable for simulating real driving operation.

[0047] In a step 101 of method 100, past data is provided. The past data represents past speed data or a past speed profile and consists of speed values, each of which is assigned to successive, defined time intervals. The defined time intervals can be constant time intervals or vary in their length. The past data can consist of a single speed value assigned to a single time interval. This single speed value can also be zero. The most recent speed value of the past speed profile is assigned to a current time interval.

[0048] In a step 102, the state vector xt is determined for the current time interval from the speed data or the past speed history. The state vector xt has as components the current speed value vt of the current time interval t, the acceleration value a t-1 of the time interval t-1 immediately before the current time interval t, a value sa,t which corresponds to the number of time intervals in which an acceleration maneuver took place immediately before the current time interval, a value se,t which corresponds to the number of time intervals in which a deceleration maneuver took place immediately before the current time interval, and a value sk,t which corresponds to the number of time intervals in which a state of constant speed persisted immediately before the current time interval.

[0049] The Fig. 1The notation sj,t used denotes the three values ​​sa,t , se,t , and sk,t , where the index j can be either a for an acceleration maneuver, e for a deceleration maneuver, or k for a state of constant velocity. The state vector can have additional components or other components corresponding to velocity values, acceleration values, changes in acceleration values, or a number of time intervals.

[0050] In a step 103, a probability value p(xt) for a current acceleration scenario and a probability value q(xt) for a current deceleration scenario are determined from the state vector using an acceleration prediction model. The probability value y for a state of constant speed is then preferably derived from the following relationship: y = 1-p(xt)-q(xt). The acceleration prediction model is preferably based on a statistical evaluation of measured driving data from a real vehicle, wherein the measured driving data consists exclusively of a temporal sequence of speed values ​​assigned to successive time intervals.

[0051] In a step 104, a random selection in the sense of a statistical random drawing of one of the three scenarios, i.e. an acceleration scenario, a deceleration scenario or the scenario of the state of a constant speed, takes place based on the probability values ​​p(xt), q(xt) and 1-p(xt)-q(xt) determined in step 103.

[0052] Using the acceleration prediction model, a probability distribution of acceleration values ​​is determined depending on the state vector for the randomly selected scenario. Preferably, a continuous probability distribution is modeled for this purpose. Furthermore, a probability can be assigned to each possible acceleration value within the randomly selected scenario.

[0053] In a step 105, a random selection in the sense of a statistical random drawing of an arbitrary acceleration value at for the current time interval t takes place from the determined probability distribution of the randomly selected scenario.

[0054] In a step 106, the randomly selected acceleration value at is integrated over the current time interval t to obtain a next predicted velocity value v t+1 for a next time interval t+1 in the future.

[0055] In step 107, the new speed value v t+1 is appended to the past speed profile. The new speed value v t+1 for the time interval t+1 is then treated as the current time interval in a second iteration of the method in step 102. By iteratively running through steps 102 to 107, a driving cycle is created which, due to the nature of the acceleration prediction model, is similar to a driving cycle measured under real conditions.

[0056] Fig. 2 shows a preferred embodiment of a computer-aided method 200 according to the invention for generating an RDE-compliant driving cycle for a vehicle.

[0057] Step 201 of method 200 is identical to step 101 of method 100 described above. Historical speed data is provided.

[0058] Step 202 of method 200 includes the above-described steps 102 and 103 of method 100. For the current time interval, the state vector xt is determined from the past speed history. Using the acceleration prediction model, a probability value p(xt) for a current acceleration scenario and a probability value q(xt) for a current deceleration scenario are determined from the state vector.

[0059] After several predicted speed values ​​have been obtained by iteratively executing method 100 and appended to the past speed history, in step 203 of method 200, the previously predicted, i.e., previously generated, speed values ​​are periodically checked for compliance with the criteria of the RDE guidelines after a defined number of iterations of method 100. For example, this periodically recurring check can be performed after a number of time intervals corresponding to the expiration of a five-minute period of the driving cycle; however, other time periods for the periodic check are also possible.

[0060] After the probability values ​​p(xt) and q(xt) for a current acceleration scenario and for a current deceleration scenario have been determined in step 202, if the check in step 203 has shown that the criteria of the RDE guidelines are not met by the already predicted speed values ​​appended to the past speed history, the determined probability values ​​p(xt) and q(xt) are corrected accordingly in a step 204.

[0061] The current acceleration or deceleration scenario thus receives corrected probability values ​​p'(xt) and q'(xt). For example, if the check in step 203 reveals that the duration of a highway journey at increased speed according to the criteria of the RDE guidelines is not met by the already predicted speed values ​​and corresponding time intervals, the probability of an acceleration scenario is increased by the correction in step 204, and the probability of a deceleration scenario is reduced accordingly.

[0062] In a step 205, one of the three scenarios, i.e. an acceleration scenario, a deceleration scenario or a scenario of the state of a constant speed, is then randomly selected based on the probability values ​​p'(xt), q'(xt) and 1- p'(xt)-q'(xt) corrected by step 204 and an arbitrary acceleration value αt is randomly selected from the probability distribution of the randomly selected scenario determined as described in the context of the method 100.

[0063] Alternatively or in addition to the correction in step 204, a correction of the randomly selected acceleration value at can be carried out in step 206 according to the check in step 203, whereby the corrected acceleration value a't is generated.

[0064] In a step 207, the corrected acceleration value a't is integrated over the current time interval t to obtain a next predicted speed value v t+1 for a next time interval t+1 in the future.

[0065] In step 208, the new speed value v t+1 is appended to the past speed profile. The new speed value v t+1 for the time interval t+1 is then treated as the current time interval in the next iteration of method 200 in step 202. By iteratively running through steps 202 to 208, a driving cycle is created which, due to the nature of the acceleration prediction model, is similar to a driving cycle measured under real conditions and complies with the RDE guidelines.

[0066] Fig. 3shows a preferred embodiment of a device 300 for generating a driving cycle for a vehicle, which is suitable for simulating real driving conditions. The device for generating the driving cycle for a vehicle has means 301 for determining a state vector of the driving cycle for a current time interval from a past speed profile.Furthermore, the device for generating the driving cycle comprises means 302 for providing an acceleration prediction model, means 303 for determining an acceleration value taking into account probabilities resulting from the acceleration prediction model and the state vector, means 304 for integrating the determined acceleration value over the current time interval in order to obtain a predicted speed value for a next time interval in the future, and means 305 for appending the predicted speed value to the past speed profile to generate the driving cycle. List of reference symbols

[0067] 300Device for generating a driving cycle for a vehicle 301Means for determining a state vector of the driving cycle 302Means for providing an acceleration prediction model 303Means for determining an acceleration value 304Means for integrating the determined acceleration value 305Means for appending the predicted speed value to the past speed profile

Claims

1. Computer-aided method (100) for generating a driving cycle for a vehicle which is suitable for simulating a driving operation, in particular a real driving operation, <b>characterized in that it comprises the following steps: determining (102) a state vector of the driving cycle for a current time interval from a past velocity progression; provision of an acceleration prediction model; determining (103, 104, 105) an acceleration value taking into account probabilities resulting from the acceleration prediction model and the state vector; integrating the determined (106) acceleration value over the current time interval to obtain a predicted velocity value for a next time interval in the future; and appending (107) the predicted velocity value to the past velocity progression to generate the driving cycle.

2. The method according to claim 1, wherein determining (103, 104, 105) an acceleration value taking into account probabilities resulting from the acceleration prediction model and the state vector comprises the following further steps: determining (103), using the acceleration prediction model as a function of the state vector, a probability value for a current scenario of acceleration and a probability value for a current scenario of deceleration and a probability value for a current scenario of a constant velocity state; and randomly selecting (104), for the current time interval, a scenario of acceleration, deceleration or constant velocity based on the probability values for current scenarios of acceleration, deceleration and constant velocity; and / or determining, by means of the acceleration prediction model as a function of the state vector, a probability distribution of acceleration values of the randomly selected scenario; and randomly selecting (105), for the current time interval, an acceleration value based on the probability distribution of acceleration values of the randomly selected scenario.

3. The method according to any one of claims 1 to 2, wherein the driving cycle is generated by iteratively executing the steps (102, 103, 104, 105, 106, 107) of the method in the listed order and the predicted velocity values are each appended to the past velocity progression from a previous iteration.

4. The method according to any one of claims 1 to 3, wherein based on the past velocity progression for the same future time intervals a plurality of predicted velocity values are obtained, so that statistical velocity distributions are obtained for future time intervals.

5. The method according to any one of claims 1 to 4, wherein the state vector for a current time interval comprises at least one of the following components: a current velocity value, one or more past velocity values, one or more acceleration values of one or more time intervals, one or more values of a change in acceleration of one or more time intervals, a value corresponding to a number of time intervals according to the duration of a currently ongoing acceleration maneuver, a value corresponding to a number of time intervals according to the duration of a currently ongoing deceleration maneuver, and a value corresponding to a number of time intervals according to the duration of a currently ongoing state of constant velocity.

6. The method according to any one of claims 1 to 5, wherein the acceleration value determined taking into account probabilities resulting from the acceleration prediction model and the state vector is based on the duration of a currently ongoing acceleration maneuver, a currently ongoing deceleration maneuver or a currently ongoing state of constant velocity.

7. The method according to any one of claims 1 to 6, wherein the current scenario of acceleration and / or the current scenario of deceleration and / or the current scenario of a state of constant velocity each have a probability distribution of acceleration values.

8. The method according to any one of claims 2 to 7, wherein an expected value of the probability distribution of acceleration values is set based on the past velocity progression.

9. The method according to any one of claims 1 to 8, wherein the acceleration prediction model is based on a statistical evaluation of measured driving data of at least one real vehicle, wherein preferably the measured driving data of the at least one real vehicle exclusively comprise a temporal sequence of velocity values.

10. Method for driving a vehicle by means of a system of adaptive driving control, in particular a driver assistance system, in particular for predictive driving functions, wherein the driving cycle of the vehicle driving ahead of the vehicle is determined by means of the method according to any one of claims 1 to 9.

11. The method according to claim 10, wherein based on the past velocity progression for the same future time intervals a plurality of predicted velocity values are obtained, so that statistical velocity distributions are obtained for future time intervals, wherein safety conditions for driving the vehicle are derived from the statistical velocity distributions.

12. A method of generating a driving cycle for a vehicle suitable for use by driver assistance systems, in particular for predictive driving functions, comprising the steps of any one of claims 1 to 9.

13. A method (200) for analyzing at least one component of a motor vehicle, wherein the at least one component or the motor vehicle is subjected to a real or simulated test operation based on at least one driving cycle which is determined by means of a method according to any one of claims 1 to 9.

14. A computer program product or computer-readable medium comprising instructions which, when executed by a computer, cause the computer to perform the steps of a method according to any one of claims 1 to 13.

15. A computer configured to perform the method steps according to any one of claims 1 to 13.